Cleaned and commented

plotting/break.R

@@ -23,7 +23,7 @@
 #
 # plotting/break.R
 #
-#
+# Plot tool useful to study date break. 
 
 
 ## 1. BREAK PLOTTING
@@ -40,24 +40,35 @@ histogram = function (data_bin, df_meta, figdir='', filedir_opt='') {
         dir.create(outdir)
     }
 
+    # Fix the major and minor date break between tick for axis
     datebreak = 10
     dateminbreak = 1
 
+    # Computes histogram by year
     res_hist = hist(data_bin, breaks='years', plot=FALSE)
+    # Gets the count by breaks
     counts = res_hist$counts
+    # In pourcentage
     counts_pct = counts/nCode * 100
+    # Gets the limits of the cells
     breaks = as.Date(res_hist$breaks)
+    # Gets the middle of the cells
     mids = as.Date(res_hist$mids)
-    
-    p = ggplot() + theme_ash +
-
-    theme(panel.grid.major.y=element_line(color='grey85', size=0.15),
-          axis.title.y=element_blank()) +
 
+    # Open a new plot with personal theme
+    p = ggplot() + theme_ash +
+        
+              # Y grid
+        theme(panel.grid.major.y=element_line(color='grey85', size=0.15),
+              # Remove y title
+              axis.title.y=element_blank()) +
+        
+        # Plot bar
         geom_bar(aes(x=mids, y=counts_pct), 
                  stat='identity',
                  fill="#00A3A8") +
-
+        
+        # X axis 
         scale_x_date(date_breaks=paste(as.character(datebreak), 
                                        'year', sep=' '),
                      date_minor_breaks=paste(as.character(dateminbreak), 
@@ -67,18 +78,19 @@ histogram = function (data_bin, df_meta, figdir='', filedir_opt='') {
                      limits=c(min(data_bin)-years(0), 
                               max(data_bin)+years(0)),
                      expand=c(0, 0)) +
-
+        
+        # Y axis
         scale_y_continuous(limits=c(0,
                                     max(counts_pct)*1.1),
                            expand=c(0, 0))
     
+    # Saving of plot
     ggsave(plot=p, 
            path=outdir,
            filename=paste('hist_break_date', '.pdf', sep=''),
            width=10, height=10, units='cm', dpi=100)
 }
 
-
 ### 1.2. Cumulative 
 cumulative = function (data_bin, df_meta, dyear=10, figdir='', filedir_opt='') {
 
@@ -92,16 +104,24 @@ cumulative = function (data_bin, df_meta, dyear=10, figdir='', filedir_opt='') {
         dir.create(outdir)
     }
 
+    # Fix the major and minor date break between tick for axis
     datebreak = 10
     dateminbreak = 1
 
+    # Computes histogram by year
     res_hist = hist(data_bin, breaks='years', plot=FALSE)
+    # Gets the count by breaks
     counts = res_hist$counts
+    # Compute the cumulative sum
     cumul = cumsum(counts)
+    # In percentage
     cumul_pct = cumul/nCode * 100
+    # Gets the limits of the cells
     breaks = as.Date(res_hist$breaks)
+    # Gets the middle of the cells
     mids = as.Date(res_hist$mids)
 
+    # Duplicates start and end value to extend graph
     mids = c(mids[1] - years(dyear), mids[1] - years(1),
                mids,
                mids[length(mids)] + years(dyear)) 
@@ -109,31 +129,44 @@ cumulative = function (data_bin, df_meta, dyear=10, figdir='', filedir_opt='') {
                   cumul_pct,
                   cumul_pct[length(cumul_pct)])
 
+    # Centers the middle date
     mids = mids + months(6)
-
+    # Shifts the breaking date to be coherent with the start
+    # of the rupture 
     breaks = breaks + 1
+    # Remove the last date because it is too much
     breaks = breaks[-length(breaks)]
 
+    # Creates a blank datebreak list to plot cumulative graph
     DB = c()
+    # For all the date break cells
     for (i in 1:length(breaks)) {
+        # Duplicates the date break for the number of times
+        # it is counts in the histogram
         DB = c(DB, rep(breaks[i], times=counts[i]))
     }
+    # Estimates the median
     q50 = as.Date(quantile(DB, probs=0.5)) + years(1)
-    
+    # Print the median
     print(paste('mediane :', q50))
-    
-    p = ggplot() + theme_ash +
-
-    theme(panel.grid.major.y=element_line(color='grey85', size=0.15),
-          axis.title.y=element_blank()) +
 
+    # Open a new plot with personal theme
+    p = ggplot() + theme_ash +
+        
+              # Y grid
+        theme(panel.grid.major.y=element_line(color='grey85', size=0.15),
+              # Remove y title
+              axis.title.y=element_blank()) +
+        
+        # Plot line of cumulative sum
         geom_line(aes(x=mids, y=cumul_pct), 
                   color="#00A3A8") +
-    
+        # Plot the median line
         geom_line(aes(x=c(q50, q50), y=c(0, 100)), 
                   color="wheat", 
                   lty='dashed') +
-
+        
+        # X axis
         scale_x_date(date_breaks=paste(as.character(datebreak), 
                                        'year', sep=' '),
                      date_minor_breaks=paste(as.character(dateminbreak), 
@@ -143,19 +176,14 @@ cumulative = function (data_bin, df_meta, dyear=10, figdir='', filedir_opt='') {
                      limits=c(min(mids)-years(0), 
                               max(mids)+years(0)),
                      expand=c(0, 0)) +
-
+        
+        # Y axis
         scale_y_continuous(limits=c(-1, 101),
                            expand=c(0, 0))
-    
+
+    # Saving plot
     ggsave(plot=p, 
            path=outdir,
            filename=paste('cumul_break_date', '.pdf', sep=''),
            width=10, height=10, units='cm', dpi=100)
 }
-
-
-
-
-
-
-